You Can Always Make New Friends: Baylor’s Defense in the Scott Drew Era

Baylor was picked 9th in the Big 12 poll and received zero votes in the preseason top 25 poll. The expectations for the Bears are low outside of Waco.

One major reason people seem down on Baylor is because of roster turnover. Jo Lual-Acuil, Terry Maston, Nuni Omot and Manu Lecomte are gone. Those four were major contributors last season, and it’s easy to wonder: how’s Baylor going to replace those players?

The concern about turnover is especially pronounced on defense. Baylor runs complicated defenses. The team switches between a 1-3-1 zone and man-to-man. King McClure told me this summer that, “The zone can take a while to learn. I feel like there are things that I’ve just now gotten down.”

Baylor also defends ball screens in a host of ways. They usually ice and drop in pick-and-roll coverage, but the Bears hard hedge and switch those actions too. There’s a lot to learn for anyone running the system.

Despite all the theories that Baylor needs continuity to be good on defense, the data shows that Baylor has been better defensively with more roster turnover:

In the graph, defensive efficiency is the national ranking for the Bears and continuity is the national percentage of minutes returning (both KenPom stats). A lower defensive efficiency number means the team gives up fewer points per possession and a lower minutes continuity number means that Baylor returned more minutes. So in 2008, Baylor was 11th in the country in returning minutes/continuity but only 110th in defensive efficiency.

Since Baylor made the NCAA Tournament in 2008, the Bears have been better on defense with more roster turnover. The relationship is statistically significant at a p value of less than .01.

Correlation does not always equal causation. That’s an important statistical principle and matters here. The 2010 Bears were much better on defense than the 2009 Bears because adding Ekpe Udoh to a defense is worth much more than the benefit of having Curtis Jerrells’ knowledge of Baylor’s defense. The 2011 Bears would have been better defensively had Udoh stayed for his senior season. There isn’t a bonus for having less continuity; it just turns out that continuity hasn’t provided much of a boost for the Bears.

2015, 2016 and 2017 provide an interesting three year window. In 2015, Baylor improved by 44 spots in defensive efficiency, despite their continuity ranking dropping by 104 spots. Baylor seemed to be much better with Kenny Chery, Lester Medford, Taurean Prince and Royce O’Neale than they were with the 2014 cast. In 2016, the team returned a lot, but that team lost O’Neale and Chery. Those two defenders were more valuable than returning so many other guys. And in 2017, Baylor benefited from Manu Lecomte taking over at the point and Lual-Acuil playing center. Those two were better defensive replacements than whatever Baylor gained from continuity. They played fewer minutes in Baylor’s system than the returning players in 2016, but their ability overwhelmed being new in Baylor’s system.

Baylor might benefit from their new guys being older. The Bears have become Transfer U, and this year they add Makai Mason and Mario Kegler to the starting lineup. Those two are entering their fourth and third years out of high school, respectively. They’re grown men and have practiced and played in college systems. The learning curve for adjusting to the speed of the game isn’t the same for them as it is for high school guys. Add in a college staff with tons of assistants and coordinators, and players have plenty of people to help them learn a new system.

What the data really shows is that talent continues to control so much in basketball. Kentucky built the best defense of the KenPom era in 2015 with a defense that played four freshmen and three sophomores in their nine man rotation. Baylor’s built their best defenses with strong defenders like O’Neale, Udoh and Lecomte. They performed exceptionally well as new players in the system. Hopefully that happens again with so much turnover.